Some useful scripts to sysadmins/dbas.
Read INSTALL to install toolkit.
- sys-genpass
- sys-nton
- sys-pubip
- sys-align
- sys-ascii
- sys-disk-error
- sys-easyhosts
- sys-ipton
- sys-lock-run
- sys-logtail
- sys-swap
- sys-repeat
- sys-daemon
- sys-checkport
- sys-echo-stderr
- sys-php-chroot
- sys-nginx-subdir
- sys-murmur3
- sys-needrestart
- sys-numa-maps
- sys-tcp-port-proxy
- sys-shell-type
- sys-sift
- sys-stalk
- sys-summary
- sys-glusterfs-rm
- sys-google-totp
- sys-traffic-capture
- sys-diskstats
- sys-http-code
- sys-http-stat
- sys-qtunnel-manage
- sys-redis-rdb-backup
- sys-dns-response-time
- sys-kill-close-wait
- sys-unmap-file
- sys-mapstat-file
- sys-mem-eat
- sys-memory-maps
- sys-memcached-check
- sys-memcached-capture
- sys-mysql-summary
- sys-mysql-adj
- sys-mysql-kill-blocked-thread
- sys-mysql-kill-long-query
- sys-mysql-block-account
- sys-mysql-health
- sys-mysql-createdb
- sys-mysql-purge-binlog
- sys-mysql-qps
- sys-mysql-repl
- sys-mysql-set-maxconnections
- sys-mysql-sql-reject
- sys-mysql-diff
- sys-mysql-error
- sys-mysql-search
- License
type: perl
Generate complex random passwords.
Example:
$ ./sys-genpass
IEZQS_hW7=exO}JG@L
help message: ./sys-genpass --help
type: perl
Convert specified number to binary/octal/decimal/hexadecimal format.
Example:
$ ./sys-nton -n 0b1111
bin: 0b1111
oct: 017
dec: 15
hex: 0x0F
$ ./sys-nton -n 017
bin: 0b1111
oct: 017
dec: 15
hex: 0x0F
$ ./sys-nton -n 15
bin: 0b1111
oct: 017
dec: 15
hex: 0x0F
$ ./sys-nton -n 0x0F
bin: 0b1111
oct: 017
dec: 15
hex: 0x0F
help message: ./sys-nton --help
type: shell
Get my public ip address by curl command.
$ ./sys-pubip
Get public ip address:
110.89.157.218
Can be used without args, return error if execute 5 seconds.
type: perl
Display ascii table, output is the same as the command man 7 ascii
$ ./sys-ascii
...
The following table contains the 128 ASCII characters.
C program '\X' escapes are noted.
Oct Dec Hex Char Oct Dec Hex Char
------------------------------------------------------------------------
000 0 00 NUL '\0' 100 64 40 @
001 1 01 SOH (start of heading) 101 65 41 A
002 2 02 STX (start of text) 102 66 42 B
003 3 03 ETX (end of text) 103 67 43 C
004 4 04 EOT (end of transmission) 104 68 44 D
005 5 05 ENQ (enquiry) 105 69 45 E
006 6 06 ACK (acknowledge) 106 70 46 F
007 7 07 BEL '\a' (bell) 107 71 47 G
010 8 08 BS '\b' (backspace) 110 72 48 H
011 9 09 HT '\t' (horizontal tab) 111 73 49 I
012 10 0A LF '\n' (new line) 112 74 4A J
013 11 0B VT '\v' (vertical tab) 113 75 4B K
...
Can be used without args.
type: perl
Get MySQL summary info.
DBI
DBD::mysql
perl-TermReadKey (if enable askpass option)
$ ./sys-mysql-summary --host 127.0.0.1 --port 19681 --user=root --askpass
Enter password :
+-127.0.0.1
version 5.6.15-rel63.0-log
server_id 101
has_gtid 1
binlog_enable 1
total_size 129.92MB
filter
binlog_format STATEMENT
max_packet 4MB
read_only 0
is_slave Yes
repl_check OK
help message: $ ./sys-mysql-summary --help
type: perl
check a host port is wether opened or not
$ ./sys-checkport -h 127.0.0.1 -p 22
port 22: open
$ ./sys-checkport -h 127.0.0.1 -p 22 -v
port 22: open
source info 127.0.0.1:53659
$ ./perl sys-checkport -h 127.0.0.1 -p 2000
port 2000: closed
$ ./sys-checkport -h 127.0.0.1 -p 2000 -u
port 2000: open | filtered -- udp port has no response
help message: $ ./sys-checkport --help
type: shell
prevent system killing MySQL process by oom-killer
$ ./sys-mysql-adj
oom_adj mysqld process
pid 31265 oom_adj = -17
pid 25653 oom_adj = -17
pid 20247 oom_adj = -17
$ ./sys-mysql-adj
no MySQL process
Can be used without args.
type: perl
check memcached whether is alive or not
set/get a memcached key to check memcached whether is alive or not, this can check a memcached even it's hang or timeout.
./sys-memcached-check -h 10.0.21.5 -p 11211
memcached 10.0.21.5:11211 check ok!
./sys-memcached-check -6 -h fe80::222:19ff:fe54:33c5%em1 -p 11211
memcached fe80::222:19ff:fe54:33c5%em1:11211 check ok!
help message: $ ./sys-memcached-check --help
type: perl
DBI
DBD::mysql
perl-TermReadKey (if enable askpass option)
kill the mysql blocking thread id if blocking times great than threthold(default is 10 times).
$ ./sys-mysql-killblockthread -h 127.0.0.1 -P 3306 -u root --askpass -v -k
Enter password :
connect to 127.0.0.1, 3306, root, xxxxxxxx ...
no thread_id blocking.
sys-mysql-killblockthread -h 127.0.0.1 -P 3306 -u root --askpass -v
Enter password :
connect to 127.0.0.1, 3306, root, xxxxxxxx ...
waiting_thread_id: 4
waiting_query: delete from t1 where a = 1
waiting_time: 5
blocking_host: 10.0.21.5:53150
blocking_thread_id: 18
blocking_query: null
kill 100123 ok.
kill 100251 ok.
help message: $ ./sys-mysql-killblockthread --help
type: shell
identifi the type of shell we used in current shell.
$ ./sys-shell-type
bash
Can be used without args
type: perl
DBI
DBD::mysql
perl-TermReadKey (if enable askpass option)
kill the mysql long query's thread if query time greater than specified time value;
note: This script will kill running sql, some update/select statements will be killed, connection db is null , state in 'Binlog Dump', 'Connect' and 'Sleep' is ignored.
$ ./sys-mysql-killlongquery -h 127.0.0.1 -u root --askpass -v -k -t 5
Enter password :
connect to 127.0.0.1, 3306, root, xxxxxxxx ...
[INFO] thread: 92, user: test@127.0.0.1:45775, db: test, state: User sleep, time: 5, query: select sleep(15)
kill 92 ok
help message: $ ./sys-mysql-killlongquery --help
type: shell
php chroot initialization work, covering DNS , php iconv and and ssl(https).
$ ./sys-php-chroot /web/php/chroot
tar: Removing leading `/' from member names
...
chroot directory /web/php/chroot initialized as
162 0 drwxr-xr-x 7 root root 71 Mar 11 14:58 /web/php/chroot
...
Calling ssl secured sites via curl in an chroot env, some /usr/lib64 library was needed:
http://jameskirsop.com/calling-ssl-secured-sites-via-curl-in-an-apache-chroot/
help message: Usage: ./sys-php-chroot /path/to/chrootdir
type: perl
Transformation of ipv4 to/from number.
$ ./sys-ipton
either ipv4 or number should be specified.
$ ./sys-ipton -i 127.0.0.1
ipv4 127.0.0.1 convert to : 2130706433
$ ./sys-ipton -n 2130706433
2130706433 convert to: 127.0.0.1
$ ./sys-ipton -i 127.0.0.1 -n 2130706685
both ipv4 address and number is not allowed
help message: Usage: ./sys-ipton --help
type: perl
block or release MySQL user account to disable/enable them login.
DBI
DBD::mysql
perl-TermReadKey (if enable askpass option)
The connect user must have create user or super privileges, the blocked account can't login when this script reverse account password.
If use kill option, the process relatate to the account will be killed.
this will set sql_log_bin off if you enable queit option, this can be useful when you don't want the password changed replicate to slave.
block account with dry-run:
$ ./sys-mysql-block-account --host 10.0.15.5 --user=root --askpass --account_user='spider' --account_host='10.0.15.%' --verbose --block --kill --dry-run
Enter password :
connect to 10.0.15.5, 3306, root, xxxxxxxx ...
[dry-run] KILL thread 3867 for spider@10.0.15.7
[dry-run] SET PASSWORD FOR spider@10.0.15.% = 19E32FB52598AF5B642ACEB857CFC344F27D1664*
block account:
$ ./sys-mysql-block-account --host 10.0.15.5 --user=root --askpass --account_user='spider' --account_host='10.0.15.%' --verbose --block --kill
Enter password :
connect to 10.0.15.5, 3306, root, xxxxxxxx ...
kill 1 connections for spider@10.0.15.7
SET PASSWORD FOR spider@10.0.15.% ok
Then the spider user password was reversed in MySQL:
+----------------------------------------------------------------------------------------------------------------+
| Grants for spider@10.0.15.% |
+----------------------------------------------------------------------------------------------------------------+
| GRANT USAGE ON *.* TO 'spider'@'10.0.15.%' IDENTIFIED BY PASSWORD '*9E32FB52598AF5B642ACEB857CFC344F27D166D3' |
| GRANT SELECT, INSERT, UPDATE, DELETE ON `spider`.* TO 'spider'@'10.0.15.%' |
+----------------------------------------------------------------------------------------------------------------+
re-execute block, the following message occured:
[block] spider@10.0.15.% already blocked
$ ./sys-mysql-block-account --host 10.0.15.5 --user=root --askpass --account_user='spider' --account_host='10.0.15.%' --verbose --release --kill
Enter password :
connect to 10.0.15.5, 3306, root, xxxxxxxx ...
SET PASSWORD FOR spider@10.0.15.% ok
spider user password is normal:
+----------------------------------------------------------------------------------------------------------------+
| Grants for spider@10.0.15.% |
+----------------------------------------------------------------------------------------------------------------+
| GRANT USAGE ON *.* TO 'spider'@'10.0.15.%' IDENTIFIED BY PASSWORD '*4661D72F443CFC758BECA246B5FA89525BF23E91' |
| GRANT SELECT, INSERT, UPDATE, DELETE ON `spider`.* TO 'spider'@'10.0.15.%' |
+----------------------------------------------------------------------------------------------------------------+
help message: Usage: ./sys-mysql-block-account --help
type: shell
bash script or command can only run once at any time by using flock
1. run command in session A
$ ./sys-lock-run -f /var/run/lockrun sleep 20
2. run command in session B
# ./sys-lock-run -f /var/run/lockrun sleep 20
2016_03_23_12_44_13 [warn] Only once can run at one time - sleep 20
note: as use flock, this tool will create lock file in /tmp dirs and delete lock file when exit.
type: shell
execute one command repeatedly until it failure.
$ ./sys-repeat ls -hl sys-checkport
-rwxr-xr-x. 1 root root 1.9K Feb 29 17:10 sys-checkport
-rwxr-xr-x. 1 root root 1.9K Feb 29 17:10 sys-checkport
-rwxr-xr-x. 1 root root 1.9K Feb 29 17:10 sys-checkport
-rwxr-xr-x. 1 root root 1.9K Feb 29 17:10 sys-checkport
-rwxr-xr-x. 1 root root 1.9K Feb 29 17:10 sys-checkport
-rwxr-xr-x. 1 root root 1.9K Feb 29 17:10 sys-checkport
...
...
help message: ./sys-repeat command ...
type: c
A simple TCP port forwarder. (fork from tcptunnel)
$ gcc -o sys-tcp-port-proxy sys-tcp-port-proxy.c
$ ./sys-tcp-port-proxy --help
Usage: ./sys-tcp-port-proxy [options]
Options:
--version
--help
--local-port=PORT local port
--remote-port=PORT remote port
--remote-host=HOST remote host
--bind-address=IP bind address
--buffer-size=BYTES buffer size
--log
--daemon
--stay-alive
1. in session A, start a proxy:
$ ./sys-tcp-port-proxy --local-port=13306 --remote-port=3306 --remote-host=127.0.0.1 --fork
2. in session B, login with mysql:
$ mysql -h 127.0.0.1 -P 13306 -uroot -p
3. session A output the connection:
tcp proxy > 2016-03-30 15:07:02: request from 127.0.0.1
type: perl
create mysql database with account user.
Connect user should be have create, create user, grant option and show databases privileges.
DBI
DBD::mysql
perl-TermReadKey (if enable askpass option)
Check replication rule:
$ perl sys-mysql-createdb --host 127.0.0.1 --user root --database mysql --askpass --createdb=test3 --account-user=user_test1 --account-host='10.0.21.%'
Enter password :
ERROR:
The mysql is in the Binlog_Ignore_DB: mysql,test,information_schema,performance_schema
maybe missing the slave update. specify the option --database value that is not in Binlog_Ignore_DB list
+-- database list:
information_schema
cztest
dashboard
db1
...
...
Check user login privileges, grant maybe skip when login user has no privileges, default value of account-pass is random charaters
$ perl sys-mysql-createdb --host 127.0.0.1 --user root --database test1 --askpass --createdb=test3 --account-user=user_test1 --account-host='10.0.21.%' --priv="select, insert" --verbose
Enter password :
connect to 127.0.0.1, 3306, root, xxxxxxxx ...
SQL: create database if not exists test3
SQL: CREATE USER user_test1@'10.0.21.%' IDENTIFIED BY 'XmhdGEZZ1KN{_fW0g3U}'
GRANT select, insert ON test3.* TO user_test1@'10.0.21.%'
[WARN] login user root@127.0.0.1 has no Grant_priv, skip ..
+-- Execute manually: GRANT select, insert ON test3.* TO user_test1@'10.0.21.%'
create database and user:
host: 127.0.0.1
port: 3306
db : test3
user: user_test1
pass: XmhdGEZZ1KN{_fW0g3U}
Create database and user ok.
$ perl sys-mysql-createdb --host 127.0.0.1 --user root --database test1 --askpass --createdb=test3 --account-user=user_test1 --account-host='10.0.21.%' --priv="select, insert" --verbose
Enter password :
connect to 127.0.0.1, 3306, root, xxxxxxxx ...
SQL: create database if not exists test3
SQL: CREATE USER user_test1@'10.0.21.%' IDENTIFIED BY 'VyhQWQCeaKxZJrroeNVP'
GRANT select, insert ON test3.* TO user_test1@'10.0.21.%'
create database and user:
host: 127.0.0.1
port: 3306
db : test3
user: user_test1
pass: VyhQWQCeaKxZJrroeNVP
The user user_test1@10.0.21.% connect ok
$ perl sys-mysql-createdb --host 127.0.0.1 --user root --database test1 --askpass --createdb=test3 --account-user=user_test1 --account-host='10.0.21.%' --priv="select, insert" --verbose
Enter password :
Already exist database: test3
help message: perl sys-mysql-createdb --help
type: shell
echo message to STDERR.
$ ./sys-echo-stderr hello world >1.txt
hello world
type: perl
Get MySQL status, qps and conn
DBI
DBD::mysql
perl-TermReadKey (if enable askpass option)
$ ./perl sys-mysql-qps -h 127.0.0.1 -u root -P 3306 --askpass -i 1
Enter password :
| -- QPS -- | -- Innodb Rows Status -- | -- Threads -- | -- Bytes --
addr time | ins upd del sel qps | ins upd del read | run conn cre | recv send
127.0.0.1:3306 16:47:28 | 0 110 0 891 1120 | 0 0 2 0 | 1 0 0 | 132044 bytes 2078183 bytes
127.0.0.1:3306 16:47:29 | 0 131 0 956 1244 | 0 0 0 0 | 4 1 0 | 144377 bytes 2226475 bytes
127.0.0.1:3306 16:47:30 | 0 142 0 996 1273 | 0 0 0 0 | 0 0 0 | 143185 bytes 2302739 bytes
127.0.0.1:3306 16:47:31 | 0 135 0 907 1179 | 0 0 1 0 | 0 0 0 | 134619 bytes 2241287 bytes
127.0.0.1:3306 16:47:32 | 0 133 0 952 1223 | 0 0 0 0 | 0 642 0 | 141494 bytes 2133514 bytes
127.0.0.1:3306 16:47:33 | 0 140 0 935 1208 | 0 0 0 0 | 0 2 0 | 131543 bytes 1962872 bytes
127.0.0.1:3306 16:47:34 | 0 133 0 1043 1318 | 0 0 1 0 | 0 1 0 | 153238 bytes 2253878 bytes
127.0.0.1:3306 16:47:35 | 0 147 0 967 1238 | 0 0 0 0 | 0 0 0 | 129724 bytes 2026643 bytes
127.0.0.1:3306 16:47:36 | 0 117 0 833 1056 | 0 0 1 0 | 1 0 0 | 121172 bytes 1974676 bytes
help message: perl sys-mysql-qps --help
type shell
Set local MySQL's max_connections to avoid error "Too many connections.."
Maybe crash when MySQL Server is very busy.
gdb
$ mysql -p -e 'select @@max_connections'
Enter password:
+-------------------+
| @@max_connections |
+-------------------+
| 100 |
+-------------------+
$ ./sys-mysql-set-maxconnections 3000
[New LWP 9029]
[New LWP 30058]
[New LWP 28201]
[New LWP 27571]
[New LWP 27476]
[New LWP 27474]
[New LWP 27473]
[New LWP 27472]
[New LWP 27471]
[New LWP 27470]
[New LWP 27469]
[New LWP 27466]
[New LWP 27465]
[New LWP 27464]
[New LWP 27463]
[New LWP 27462]
[New LWP 27461]
[New LWP 27460]
[New LWP 27459]
[New LWP 27458]
[New LWP 27457]
[Thread debugging using libthread_db enabled]
0x00007fa0b2371113 in poll () from /lib64/libc.so.6
set pid 27454 max_connections = 3000
$ mysql -p -e 'select @@max_connections'
Enter password:
+-------------------+
| @@max_connections |
+-------------------+
| 3000 |
+-------------------+
help message: ./sys-mysql-set-maxconnections connections_number
type: shell
See the memory mapping of a process
$ ./sys-memory-maps 25005
share = 100164 Kbytes
private = 39332 Kbytes
total = 139496 Kbytes ( 71.80 % shareable)
help message: sys-memory-maps pid
type: c
A simple tool to sniffer MySQL query and send RST to source ip and port if you want to disable the sql execute.
This tool is similar to iptables which has the string option to match the data in tcp packages.
sys-mysql-sql-reject is fork from snapper: https://github.com/vr000m/Snapper
libpcap
# gcc -g -Wall -o sys-mysql-sql-reject -lpcap sys-mysql-sql-reject.c
in session A, MySQL Server will send RST mark to client when you execute the sql "select user,host from mysql.user"
./sys-mysql-sql-reject em1 "tcp dst port 3306" "select user,host from mysql.user"
snapper - based on Sniffer example using libpcap
extended by Varun Singh / Copyright (c) 2005 The Tcpdump Group
THERE IS ABSOLUTELY NO WARRANTY FOR THIS PROGRAM.
NET: 10.0.21.0 a03fe00 CMASK: 255.255.255.0 ffffff00
Device: em1
Filter expression: tcp dst port 3306
MySQL query filter: select user,host from mysql.user
P107: 10.0.21.7 -> 10.0.21.17 TCP 64722 -> 3306 query: desc tm
P108: 10.0.21.7 -> 10.0.21.17 TCP 64722 -> 3306
P109: 10.0.21.7 -> 10.0.21.17 TCP 64722 -> 3306 query: delete from tm where a = 'hah'
P110: 10.0.21.7 -> 10.0.21.17 TCP 64722 -> 3306
P111: 10.0.21.7 -> 10.0.21.17 TCP 64722 -> 3306 query: select user,host from mysql.user
P112: 10.0.21.7 -> 10.0.21.17 TCP 64722 -> 3306
in session B:
mysql> desc tm;
+-------+------------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+-------+------------------+------+-----+---------+----------------+
| a | int(10) unsigned | NO | PRI | NULL | auto_increment |
+-------+------------------+------+-----+---------+----------------+
1 row in set (0.03 sec)
mysql> delete from tm where a = 'hah';
Query OK, 0 rows affected (0.00 sec)
mysql> select user,host from mysql.user;
ERROR 2013 (HY000): Lost connection to MySQL server during query
help message:
Usage: sys-mysql-sql-reject [interface] [tcp_filter] [sql_filter]
Options:
interface Listen on <interface> for packets.
tcp_filter PCAP Filter to apply on packets.
sql_filter MySQL query to match on packets
type: perl
see the local device disk status
read more from kernel Documentation/iostats.txt
$ ./sys-diskstats -d sda3 -i 5
iops rs rs_mer ws ws_mer rs_kB ws_kB e_iot e_iot_w r_t w_t
19:32:36 98.00 0.00 0.00 98.00 0.40 0.00 574.20 6.60 ms 6.60 ms 0.00 ms 6.60 ms
19:32:41 331.40 0.00 0.00 331.40 0.00 0.00 9550.20 19.20 ms 33.00 ms 0.00 ms 33.00 ms
19:32:46 90.60 0.00 0.00 90.60 0.80 0.00 570.20 3.00 ms 3.00 ms 0.00 ms 3.20 ms
19:32:51 125.80 0.00 0.00 125.80 0.00 0.00 3263.80 5.20 ms 14.20 ms 0.00 ms 14.20 ms
19:32:56 93.20 0.00 0.00 93.20 0.00 0.00 533.00 3.80 ms 3.80 ms 0.00 ms 3.80 ms
19:33:01 296.40 0.00 0.00 296.40 0.00 0.00 9362.20 15.80 ms 22.60 ms 0.00 ms 22.60 ms
help message: ./sys-diskstats -h
item list:
iops => io operations per second
rs => disk reads per second
rs_mer => disk read merged per second
ws => disk write per second
ws_mer => disk write merged per second
rs_kB => disk kBytes read per second
ws_kB => disk kBytes write per second
e_iot => disk io time in ms
e_iot_w => disk io time weighted
r_t => disk read time in ms
w_t => disk write time in ms
type: perl
Explains the meaning of an HTTP status code.
$ ./sys-http-code -c 100
Status code: 100
Message: Continue
Code explanation: Request received, please continue
$ ./sys-http-code -c 10x
Status code: 100
Message: Continue
Code explanation: Request received, please continue
Status code: 101
Message: Switching Protocols
Code explanation: Switching to new protocol; obey Upgrade header
Status code: 102
Message: Processing
Code explanation: WebDAV; RFC 2518, Server has received and is processing the request.
Status code: 103
Message: Processing
Code explanation: Server has received and is processing the request.
$ ./sys-http-code -c all
Status code: 100
Message: Continue
Code explanation: Request received, please continue
Status code: 101
Message: Switching Protocols
Code explanation: Switching to new protocol; obey Upgrade header
...
...
Status code: 599
Message: Network connect timeout error
Code explanation: Network connect timeout behind the proxy.
help message: ./sys-http-code -h
type: perl
managing your /etc/hosts file.
$ sys-easyhosts -m list
localhost => 127.0.0.1 (on)
localhost => ::1 (on)
localhost.localdomain => 127.0.0.1 (on)
localhost.localdomain => ::1 (on)
localhost4 => 127.0.0.1 (on)
localhost4.localdomain4 => 127.0.0.1 (on)
localhost6 => ::1 (on)
localhost6.localdomain6 => ::1 (on)
# check domain
$ sys-easyhosts -m has google.com
google.com not found in hosts file
# add domain
$ sys-easyhosts -m add google.com 172.217.8.14
ok - [add] google.com 172.217.8.14
# new list in /etc/hosts
$ cat /etc/hosts
127.0.0.1 localhost.localdomain localhost4.localdomain4 localhost4 localhost
172.217.8.14 google.com
::1 localhost6.localdomain6 localhost.localdomain localhost6 localhost
# disable doamain list
$ sys-easyhosts -m off google.com
ok - [off] google.com
$ cat /etc/hosts
127.0.0.1 localhost4.localdomain4 localhost.localdomain localhost4 localhost
# 172.217.8.14 google.com
::1 localhost6.localdomain6 localhost.localdomain localhost6 localhost
use --help option for more usage message, ipv4 and ipv6 are both support.
mode must set list, fix, add, aff, on, off, del, has. and the domain value is $ARGV[0](google.com), ip value is $ARGV[1](172.217.8.14), all mode support dryrun
that only print result, don't change hosts file.
type: perl
Measure the dns server response time.
perl-Net-DNS
centos/redhat can install bind-utils so that perl-Net-DNS lookup domain easily
$ ./sys-dns-response-time -d highdb.com
query from dns server: 114.114.114.114, elapsed ms: 31
query from dns server: 114.114.114.114, elapsed ms: 29
query from dns server: 114.114.114.114, elapsed ms: 28
help message: ./sys-dns-response-time -h
type: perl
kill the tcp close_wait state connections without restart server program.
$ netstat -tulnap|grep CLOSE | grep '10.0.21.17'
tcp 0 0 ::ffff:10.0.21.5:7200 ::ffff:10.0.21.17:55365 CLOSE_WAIT 2681/./audit_server
$ ./sys-kill-close-wait --src_host 10.0.21.17 --src_port 55365 --dst_host 10.0.21.5 --dst_port 7200
kill close_wait connection 10.0.21.17:55365 => 10.0.21.5:7200
send ack ok!
help message: ./sys-kill-close-wait -h
type: shell
capture interface packets when traffic greater than a value
tcpdump
capture when interface em1's in traffic greater than 1000000 bytes
$ bash sys-traffic-capture em1 in 1000000
2016_10_22_11_21_19 [info] capture em1 300000 packets into em1-2016_10_22_11_21_15.pcap
type: shell
get curl statistics made simple.
fork from https://github.com/b4b4r07/httpstat with no bc command need.
get https://highdb.com url statistics
sys-http-stat https://highdb.com -k
HTTP/1.1 200 OK
Server: nginx
Date: Mon, 06 Feb 2017 10:04:57 GMT
Content-Type: text/html; charset=UTF-8
Transfer-Encoding: chunked
Connection: keep-alive
Link: <https://highdb.com/wp-json/>; rel="https://api.w.org/"
website ip address: 85.90.244.138:443
Body stored in: /tmp/httpstat-body.6525266621486375037
DNS Lookup TCP Connection SSL Handshake Server Processing Content Transfer
[ 4ms | 233ms | 620ms | 2322ms | 233ms ]
| | | | |
namelookup:4ms | | | |
connect:237ms | | |
pretransfer:857ms | |
starttransfer:3179ms |
total:3412ms
speed_download 7440.4 KiB, speed_upload 0.0 KiB
type: perl
compare MySQL database schema and privilege at different time. inspired by https://github.com/aspiers/mysqldiff
mysql root@[localhost:s3306 test] > drop table table_add;
Query OK, 0 rows affected (0.02 sec)
# perl sys-mysql-diff -h 127.0.0.1 -P 3306 -u root --askpass -d test
Enter password :
DROP TABLE `table_add`;
mysql root@[localhost:s3306 test] > CREATE TABLE `table_add` (
`id` int(10) NOT NULL AUTO_INCREMENT,
`name` varchar(20) DEFAULT NULL,
`source` tinyint(1) DEFAULT NULL,
`create_time` datetime DEFAULT NULL,
`type` tinyint(1) NOT NULL DEFAULT '0',
PRIMARY KEY (`id`),
KEY `idx_cretime` (`create_time`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
Query OK, 0 rows affected (0.02 sec)
#perl sys-mysql-diff -h 127.0.0.1 -P 3306 -u root --askpass -d test
Enter password :
CREATE TABLE table_add (
id int(10) NOT NULL AUTO_INCREMENT,
name varchar(20) DEFAULT NULL,
source tinyint(1) DEFAULT NULL,
create_time datetime DEFAULT NULL,
type tinyint(1) NOT NULL DEFAULT '0',
PRIMARY KEY (id),
KEY idx_cretime (create_time)
) ENGINE=InnoDB DEFAULT CHARSET=utf8
mysql root@[localhost:s3306 test] > alter table table_add add column descrips varchar(100) not null default '';
Query OK, 0 rows affected (0.03 sec)
Records: 0 Duplicates: 0 Warnings: 0
mysql root@[localhost:s3306 test] > alter table table_add add key idx_source(`source`);
Query OK, 0 rows affected (0.01 sec)
Records: 0 Duplicates: 0 Warnings: 0
# perl sys-mysql-diff -h 127.0.0.1 -P 3306 -u root --askpass -d test
Enter password :
ALTER TABLE `table_add` ADD COLUMN `descrips` varchar(100) NOT NULL DEFAULT '';
ALTER TABLE `table_add` ADD INDEX `idx_source` (`source`);
mysql root@[localhost:s3306 test] > alter table table_add drop key idx_source;
Query OK, 0 rows affected (0.01 sec)
Records: 0 Duplicates: 0 Warnings: 0
mysql root@[localhost:s3306 test] > alter table table_add drop column descrips;
Query OK, 0 rows affected (0.02 sec)
Records: 0 Duplicates: 0 Warnings: 0
# perl sys-mysql-diff -h 127.0.0.1 -P 3306 -u root --askpass -d test
Enter password :
ALTER TABLE `table_add` DROP COLUMN `descrips`;
ALTER TABLE `table_add` DROP INDEX `idx_source`;
mysql root@[localhost:s3301 test] > set global wait_timeout = 1000;
Query OK, 0 rows affected (0.00 sec)
mysql root@[localhost:s3301 test] > revoke all on test.* from user_test@`10.3.254.%`;
Query OK, 0 rows affected (0.00 sec)
# perl sys-mysql-diff -h 127.0.0.1 -P 3301 -u root --askpass -d test -t -r
Enter password :
SET GLOBAL wait_timeout = 1000;
REVOKE ALL PRIVILEGES ON test.* FROM 'user_test'@'10.0.21.%';
mysql root@[localhost:s3301 test] > set password for user_test@`10.0.21.%` = password('xxxxxx');
Query OK, 0 rows affected (0.00 sec)
mysql root@[localhost:s3301 test] > revoke select on test.* from user_test@`10.0.21.%`;
Query OK, 0 rows affected (0.00 sec)
# perl sys-mysql-diff -h 127.0.0.1 -P 3301 -u root --askpass -d test -t -r
Enter password :
SET PASSWORD FOR 'user_test'@'10.0.21.%' = '*4661D72F443CFC758BECA246B5FA89525BF23E91';
REVOKE SELECT ON test.* FROM 'user_test'@'10.0.21.%';
help message: perl sys-mysql-diff --help
type: shell
remove glusterfs file
switch to glusterfs volum dir before you remove a file
$ sys-glusterfs-rm slave_info.log
Would remove the following
.glusterfs/9f/88/9f880673-f47f-458c-b0a5-46316f9377b0
rm: remove regular file `.glusterfs/9f/88/9f880673-f47f-458c-b0a5-46316f9377b0'? y
rm: remove regular file `slave_info.log'? y
help message: sys-glusterfs-rm --help
type: perl
run script with daemonize way.
#!/bin/bash
while(true); do
F=$(date +%s.%N)
echo $F
sleep 1
done
run t.sh with sys-daemon:
./sys-daemon -p /tmp/tt.pid -o /tmp/tt.log /tmp/t.sh
or
perl sys-daemon -p /tmp/tt.pid -o /tmp/tt.log /tmp/t.sh
Overwritting PID file /tmp/tt.pid because PID 5705 is not running.
# ps -ef|grep t.sh | grep -v 'grep'
root 5705 1 0 19:31 ? 00:00:00 /bin/bash /tmp/t.sh
use help option to read more.
type: perl
Explain MySQL error codes, like the mysql perror command.
Covers up to MySQL 5.7.14-8. All of the code msg from Percona Server 5.7.14-8.
read more from source_code/sql/share/errmsg-utf8.txt.
# perl sys-mysql-error -c 1087
MySQL error code 1087 (ER_LOAD_INFO): Records: %ld Deleted: %ld Skipped: %ld Warnings: %ld
# perl sys-mysql-error -c 4000
Illegal error code: 4000
use help option to read more.
type: shell
remote backup redis rdb file, execute bgsave before transfer rdb file.
# ./sys-redis-rdb-backup -h 10.0.21.5 -P 6380 -d /srv/redis_backup
2017_06_27_14_59_40 [info] test with ssh to 10.0.21.5 ok
-NOAUTH, maybe need redis password, use -p options
# ./sys-redis-rdb-backup -i /home/mysql/.ssh/id_rsa -h 10.0.21.5 -P 6380 -d /srv/redis_backup -p xxxxxx
2017_06_27_14_59_49 [info] test with ssh to 10.0.21.5 ok
2017_06_27_14_59_52 [info] start at 20170627145949
receiving incremental file list
redis32.rdb
sent 30 bytes received 467 bytes 994.00 bytes/sec
total size is 389 speedup is 0.78
2017_06_27_14_59_52 [info] 10.0.21.5:6380 backup to /srv/redis_backup/10.0.21.5-6380-20170627145949.rdb ok
use help option to read more.
type: perl
get google Authentication which used the Time-based One-time password Algorithm.
# ./sys-google-totp --secret FNENMTM3BTB42EKM
totp message:
secret: FNENMTM3BTB42EKM
494065 (9 second(s) remaining)
use help option to read more.
type: perl
create the subdirectory in year/month/day format at nginx log directory.
you can use sys-nginx-subdir to create the sub dirs with dynamic nginx
access_log path(the method cronolog and pipe is very inconvenient and
maybe block when pipe is no consumer; the nginx perl module can reduce
performance when you use perl module to check and create dirs).
the access_log in nginx configure:
location ~ ^/xxx/ {
if ($time_iso8601 ~ "^(\d{4})-(\d{2})-(\d{2})T(\d{2})")
{
set $year $1;
set $month $2;
set $day $3;
set $hour $4;
}
access_log /data/nginx/logs/$year/$month/$day/$hour.log;
}
create dir with sys-nginx-subdir:
# ./sys-nginx-subdir -v -d /data/nginx/logs -t 1 -m
datestring: 2017-08-12 17:30:43
will create /web/nginx/logs/2017/08/12
[ok] already exists /web/nginx/logs/2017/08/12
create ok!
then reload the nginx with -s option.
use help option to read more.
type: shell
simple swap setup script for Linux
# sys-swap 8G /web/swap/swapfile
step 1. dd /web/swap/swapfile with bs: 4096 and count: 2097152
2097152+0 records in
2097152+0 records out
8589934592 bytes (8.6 GB) copied, 66.3824 s, 129 MB/s
step 2. mkswap file /web/swap/swapfile
Setting up swapspace version 1, size = 8388604 KiB
no label, UUID=34ebfcf2-9d88-4326-8e16-a48aec26e1f1
step 3. chmod file /web/swap/swapfile to 0600
step 4. swapon file /web/swap/swapfile
step 5. add swap info to fstab
/web/swap/swapfile swap swap defaults 0 0
Done! You now have a 8 swap file at /web/swap/swapfile
show the help message with no argument.
type: perl
search mysql metadata info with REGEXP syntax, include database name, table name, column name, variable or status.
# sys-mysql-search -h 10.0.21.5 -P 3301 -u root --askpass -s '.*count|test' -m t
Enter password :
10.0.21.5:3301
SCHEMA: graph, TABLE: endpoint_counter, BASE TYPE: BASE TABLE
SCHEMA: test, TABLE: count_test, BASE TYPE: BASE TABLE
SCHEMA: test, TABLE: share_count, BASE TYPE: BASE TABLE
SCHEMA: test, TABLE: test_count, BASE TYPE: BASE TABLE
you can use SQL regexp in --search option. use the help option to read more.
type: perl
murmur Murmur3 32bit hash calculate. read more from murmurHash
# sys-murmur3 -s "hello world"
2951810136
use help option to read more.
type: perl
A simple tool for discover MySQL replication topology, also support master-master topology.
sys-mysql-repl can get the master, slave status, common info(server_id, binlog, filter ...) and check repl health, but cann't get replication info if the master and slave have different port and give the --host option with slave value.
the --host can be set both master and slave ip address:
perl sys-mysql-repl --host 10.0.21.17 --port 3303 --user monitor --askpass
Enter password :
+-10.0.21.17:3303
version 5.5.36-34.1-rel34.1-log
server_id 68839
has_gtid Not Support
tx_isolation REPEATABLE-READ
binlog_enable 1
filter binlog_ignore_db: information_schema,mysql,performance_schema,test;
binlog_format ROW
max_packet 32MB
read_only 0
+-10.0.21.7:3303
version 5.5.36-34.1-rel34.1-log
server_id 462055
has_gtid Not Support
tx_isolation REPEATABLE-READ
binlog_enable 1
filter replicate_ignore_db: information_schema,mysql,performance_schema,test;
binlog_format ROW
max_packet 32MB
read_only 1
repl_check OK
use help option to read more.
type: perl
check the mysql health, this is forked from MySQLTunner-perl, read more from mysql_health.
use help option to read more.
type: shell
check the disk media or other error count.
All the check are based on Broadcom's MegaCli command.
# ./sys-disk-error
=> disk numbers: 5
slot: 0
inquiry: SEAGATE ST9600205SS
raw_size: 572325MB
media_error: 0
other_error: 6
predictive_failure: 0
type: SAS
firmware_state: Online
slot: 2
inquiry: SEAGATE ST9600205SS
raw_size: 572325MB
media_error: 12
other_error: 6
predictive_failure: 0
type: SAS
firmware_state: Online
type: shell
Read the log file from last checkpoint, it's useful when analsis log file with a incrementally way.
$ echo 1 >/tmp/t.log
$ sys-logtail -f /tmp/t.log
1
$ echo 2 >>/tmp/t.log
$ sys-logtail -f /tmp/t.log
2
use --help option for more usage message.
type: perl
report the numa information about running processes.
# ./sys-numa-maps -p 31647
N0 : 597833 ( 2.28GB)
N1 : 581409 ( 2.22GB)
active : 1864 ( 7.28MB)
anon : 1176537 ( 4.49GB)
dirty : 1176537 ( 4.49GB)
kernelpagesize_kB : 1740 ( 6.80MB)
mapmax : 638 ( 2.49MB)
mapped : 2804 ( 10.95MB)
read more from kernel-doc/Documentation/vm/numa_memory_policy.txt, and use --help option to read usage message.
type: c
simple tool to eat system memory, and it does not release the allocated memory.
# ./mem --verbose
[verbose] allocated max memory: 100000 MB, step size: 10
[2018-09-27 19:07:06] allocated 10 MB
[2018-09-27 19:07:08] allocated 20 MB
[2018-09-27 19:07:10] allocated 30 MB
[2018-09-27 19:07:12] allocated 40 MB
use --help option to read usage message.
type: shell
manage the qtunnel by the configure file.
$ sys-qtunnel-manage -l
All tags in config file: /etc/conn.conf:
server1 - mode: server; 10.0.21.5:16380 --> 10.0.21.7:6380
server2 - mode: server; 10.0.21.5:13310 --> 10.0.21.7:3310
$ sys-qtunnel-manage -r
Already running tags:
10844 -> server1
10898 -> server2
$ sys-qtunnel-manage -t server1 -k
2018_09_01_09_42_56 [info] kill the qtunnel with tag server1 ok
$ sys-qtunnel-manage -r
Already running tags:
10898 -> server2
use -h option for more usage message.
type: shell
capture memcached traffic and parsed to get item list.
$ sys-memcached-capture -p 11212 -i eth0 -e 'src 10.0.21.5'
- start tcpdump ...
- use 'Crtl+c' to stop!
tcpdump: listening on any, link-type LINUX_SLL (Linux cooked), capture size 65535 bytes
^C5 packets captured
10 packets received by filter
0 packets dropped by kernel
- SIGINT received, kill tcpdum...
- parse the traffic file /tmp/2018_12_11_12_07_22-11212-memcached.pcap...
2018-12-10 12:02:58.950259 - get tags
2018-12-10 12:03:04.910211 - set msg 0 0 3
2018-12-10 12:03:10.103213 - get msg
2018-12-10 12:03:13.430215 - delete msg
-e is optional, use -h option for more usage message.
type: perl
aligns output from other tools to columns.
# vmstat | tail -2
r b swpd free buff cache si so bi bo in cs us sy id wa st
0 0 0 5588044 0 2459224 0 0 0 1 0 0 0 0 100 0 0
# vmstat | tail -2 | sys-align
r b swpd free buff cache si so bi bo in cs us sy id wa st
1 0 0 5585928 0 2459304 0 0 0 1 0 0 0 0 100 0 0
use --help option for more usage message.
type: Bash
Browses files created by sys-stalk.
fork from pt-sift. but add the following features:
1. remove all about MySQL summary info;
2. add report on tcpdump, tcprstat, perf, disk io, httpstat, netstat.. etc;
3. remove the pt-pmap, pt-diskstats ... dependencies;
use --help option for more usage message.
type: Bash
Collect forensic data about system when problems occur.
fork from pt-stalk. but add the following features:
1. remove all about MySQL summary info;
2. add tcpdump, tcprstat, numactl, ss ... commands;
3. add perf support;
use --help option for more usage message.
type: Bash
Summarize system information nicely.
fork from pt-summary, but add the following features:
1. fix the file command read symblinks file error;
2. ignore /lib/libc.so.6 error when load snoopy library;
3. fix fio-status output parse error;
4. fix partition error when sort and join doesn't match;
5. fix process list parse error;
6. add Centos7/RHEL7 support when parse free command output;
7. MegaCli support MegaCli and MegaCli64;
8. add zpool/zfs report;
9. fix top command output;
use --help option for more usage message.
type: Perl
purge MySQL binary logs, you can retain binlog files by count or hours type. the option retain is integer value, default is 5, it retain number files if type is count, retain about number hours files if type is hour.
the connect user must have super, process, replication_client privileges. option --host can be set both master and slave ip address, --host and --port option must be slave info if master and slave have different port:
$ sys-mysql-purge-binlog --host 10.0.21.5 --port 3301 --user monitor --askpass --purge --type count --retain 2
Enter password :
2019-05-09T18:43:11 [purge check] mysql replication status:
+-10.0.21.5:3301
version 5.6.38-83.0-log
server_id 1379557
has_gtid 0
binlog_enable 1
tx_isolation REPEATABLE-READ
binlog_format MIXED
charset utf8
max_packet 1024MB
read_only 0
+-10.0.21.7:3301
version 5.6.38-83.0-log
server_id 1445093
has_gtid 0
binlog_enable 1
tx_isolation REPEATABLE-READ
binlog_format MIXED
relay_master_file mysql-bin.000083
relay_behind_second 0
charset utf8
max_packet 1024MB
read_only 0
repl_check OK
2019-05-09T18:43:11 retain 2 binary logs is approximate to purge before mysql-bin.000081
2019-05-09T18:43:11 purge binary logs to mysql-bin.000081 ok!
$ sys-mysql-purge-binlog --host 10.0.21.5 --port 3301 --user monitor --askpass --purge --type hour --retain 3
Enter password :
2019-05-09T18:43:25 [purge check] mysql replication status:
+-10.0.21.5:3301
version 5.6.38-83.0-log
server_id 1379557
has_gtid 0
binlog_enable 1
tx_isolation REPEATABLE-READ
binlog_format MIXED
charset utf8
max_packet 1024MB
read_only 0
+-10.0.21.7:3301
version 5.6.38-83.0-log
server_id 1445093
has_gtid 0
binlog_enable 1
tx_isolation REPEATABLE-READ
binlog_format MIXED
relay_master_file mysql-bin.000083
relay_behind_second 0
charset utf8
max_packet 1024MB
read_only 0
repl_check OK
2019-05-09T18:43:25 master generate 0 binlog files at per hours
use --help option for more usage message.
type: c
unmap a file from file system cache, read more from unmap_mysql_logs.
$ free -m
total used free shared buff/cache available
Mem: 64216 22761 3188 3210 38266 37630
$ sys-unmap-file -f /var/lib/mysql/data/mysql-bin.000087 -n 90 -s
$ free -m
total used free shared buff/cache available
Mem: 64216 22760 3937 3210 37518 37631
use -h option to read more usage.
type: c
get cache statistics for a file. read more from nocache.
$ sys-mapstat-file -f MySQL-Linux.x86_64.ssl101.tar.gz
page in cache: 16791/16791 (100.0%) [filesize=67161.5K, pagesize=4K]
use -h option to read more usage.
type: perl
check and report the need restart services, does not support auto restart sevice.
note: only useful when Linux is running in systemd service. almost of the check method is reference from needrestart.
# sys-needrestart -p 1841
pid(1841) - [systemd-machined.service] is need restart.
# sys-needrestart
The following service should restart as it
uses deleted or non-existing binary file.
pid service
1 - systemd manager
526 - systemd-journald.service
548 - lvm2-lvmetad.service
560 - systemd-udevd.service
765 - NetworkManager.service
766 - polkit.service
772 - irqbalance.service
775 - systemd-logind.service
1082 - tuned.service
use --help option for more usage message.
MIT / BSD